A Mutual Information Based on Ant Colony Optimization Method to Feature Selection for Categorical Data Clustering

Z. Shojaee; S. A. Shahzadeh Fazeli; E. AbbasiF. AdibniaF. MasuliS. Rovetta

首页> 外文期刊>Iranian Journal of Science >A Mutual Information Based on Ant Colony Optimization Method to Feature Selection for Categorical Data Clustering

【24h】

A Mutual Information Based on Ant Colony Optimization Method to Feature Selection for Categorical Data Clustering

机译：A Mutual Information Based on Ant Colony Optimization Method to Feature Selection for Categorical Data Clustering

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

By improving feature extraction techniques, high-dimensional datasets emerge more frequently, in which irrelevant or redundant features may appear. This curse of dimensions will affect the performance of clustering algorithms. Motivated by this, there is an increasingly necessity of selecting the most informative features. On the other hand, high time complexity of brute force methods makes heuristic techniques a better substitute. In this paper, we apply a mutual information based on ant colony optimization technique to select the most informative features by omitting redundant and irrelevant features. In our proposed method, MIMRFS (Maximum Informativity Minimum Redundancy Feature Selection), each ant individually decides to choose a subset of best features. Then the attitude of all ants is gathered and the final selected subset is chosen. This selected subset will be used to cluster whole instances. Experimental results show the high performance of MIMRFS. According to empirical results, the accuracy of MIMRFS is 96%, 83%, and 85% on three datasets Engine, Chess, and Lymphography which have an increase in comparison with other categorical clustering techniques.

著录项

来源
《Iranian Journal of Science》 |2023年第1期|175-186|共12页
作者
Z. Shojaee; S. A. Shahzadeh Fazeli; E. AbbasiF. AdibniaF. MasuliS. Rovetta;
展开▼
作者单位

Department of Computer Science, Yazd University, Yazd, Iran,Parallel Processing Laboratory, Yazd University, Yazd, Iran;

Department of Computer Science, Yazd University, Yazd, Iran;

Computer Engineering Department. Yazd University, Yazd, IranDIBRIS, University of Genoa, Genoa, Italy;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类
关键词
Feature selection; Clustering; Categorical; Ant colony optimization;

A Mutual Information Based on Ant Colony Optimization Method to Feature Selection for Categorical Data Clustering

摘要

著录项

相关主题

期刊订阅